Outline

  1. digital revolution or hype?
  2. about us
  3. goals of this course

AI: A non-standard introduction

The world has changed, hasn’t it?

The era of AI:
Big data, big government, big business

… and big empowerment

Group discussion

What makes a device looking intelligent?

AI is a moving target with respect to …

  • human capabilities
  • technological abilities

Transfer of human intelligence 🤖

from static machines to more flexible devices

  • mimicking intelligent behavior
    • perception: reading + seeing + hearing
    • generation: speaking + writing + drawing
    • moving in the physical world
  • flexibility and contextual adaptability
  • reproducing any media form

Seeing like a human?

An image segmentation by Facebook’s Detectron2 (Wu et al. 2019)

Hearing and speaking like a human?

Simulating (multiple) humans?

Beyond linear conversations

Outsmarting humans?

Debunk some myths around ChatGPT

  • is a brand, large-language models (LLM) is the technology

  • generates fluent text, not necessarily truthful

  • is highly useful, although it understands little

  • what is tough for humans might be easy for a model; and vice-versa

  • is English-focused, multi-linguality is limited

  • generates non-reproducible outputs

  • generated text cannot be detected (except verbatim parts)

  • yesterday’s version might be different from today’s

  • open-source is chasing OpenAI and the other Sillicon Valley giants

Where does the smartness come from?

Number of words exposed

An LLM is amazing but …

… it is also a stochastic parrot. 🦜

(Bender et al. 2021)

The LLM apocalypse

A post-apocalyptic take on education (Mollick 2023)

  • chat bots are another tool like Google Search
  • chat bots are your personal tutor
    • productivity tool to become faster and better
    • helping you to think, not replacing thinking
    • don’t trust blindly
    • be transparent

LLMs are a tool, learn how to use it 👍

These people do not exist

Generated Images by a neural network (Karras et al. 2020)

Faces generated by StyleGAN. Generate more faces!

Multimodality and guidance

Guided generation of text, audio, images, video

Prompt: Elephant amigurumi walking in savanna, a professional photograph, blurry background

State-of-the-Art image generation model Imagen3. (Imagen-Team-Google et al. 2024)
Use on Gemini platform.

Adapt images using text prompts

Editing pictures with Muse using natural language (Chang et al. 2023)

Erase or edit reality

For your Instagram or fake politics

Modify pictures thoroughly in Google Photos

From image to video generation 🎥

Synthesize any content with ever increasing quality

Performance Google Veo 2 vs OpenAI Sora (end vs start 2024)

Real-time, multimodal interaction

Fusing the digital and the physical world 👀👂🧠

Interact with Google’s Gemini using text, voice, video, or screen sharing

Artificial Intelligence (AI)

(Converging) Subfields

  • Natural Language Processing (NLP)
  • Computer Vision (CV)
  • Robotics 🤖

How does computer intelligence work?

  • concepts with overlapping meaning
    • Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL)
  • learn patterns from lots of data
    • more recycling than genuine intelligence
    • theory agnostically
  • supervised training is the most popular
    • learn relation between input and output

AI is also hype 📣

AI = from humankind import solution

Looking beyond the technology

  • smarter: ever more powerful

  • cheaper: intelligence is getting commoditized

  • democratized (?): open vs closed models

  • competitive: geopolitical rivalry

  • unecological: raising ecological costs

Why this matters for
Social Science

Computational Social Science

data-driven research

Group discussion

What kind of data is there?

What data is relevant for social science?

  • data as traces of social behaviour
    • tabular, text, image
  • datafication
    • sensors of smartphone, digital communication
  • much of human knowledge compiled as text

About the mystery of coding

coding is like…

  • cooking with recipes
  • superpowers

… to tackle complex problems on scale 🪄

Everyone can aquire coding powers!

About us

Goals of this course 🎯

What you learn

  • collect and curate data
  • computationally analyze, interpret, and visualize texts
  • digital literacy + scholarship
  • problem-solving capacity

Learnings from previous courses

  • too much content, too little practice
  • programming can be overwhelming
  • learning by doing, doing by googling (ChatGPT?!)

Levels of proficiency

  1. awareness of today’s computational potential
  2. analyzing existing datasets
  3. creating + analyzing new datasets
  4. applying advanced machine learning

How I teach

  • computational practice
  • critical perspective on technology
  • lecture-style introductions
  • hands-on coding sessions
  • discussions + experiments in groups

Provisional schedule

Date Topic
20 February 2025 Introduction + Where is the digital revolution?
27 February 2025 no lecture (Fasnacht)
06 March 2025 Text as Data
13 March 2025 Setting up your Development Environment
20 March 2025 Introduction to the Command-line
27 March 2025 Basic NLP with Command-line
03 April 2025 Introduction to Python in VS Code
10 April 2025 Working with (your own) Data
17 April 2025 Data Analysis of Swiss Media
24 April 2025 no lecture (Osterpause)
01 May 2025 Ethics and the Evolution of NLP
08 May 2025 NLP with Python
15 May 2025 NLP with Python II + Working Session
22 May 2025 Mini-Project Presentations + Discussion
29 May 2025 no lecture (Christi Himmelfahrt)

TL;DR 🚀

You will be tech-savvy…

…yet no programmer applying fancy machine learning

Requirements

  • no technical skills required ✅
    • self-contained course
  • laptop (macOS, Win11, Linux) 💻
    • update system
    • free up at least 15GB storage
    • backup files

Grading ✍️

  • 2 assignments during semester
    • no grades (pass/fail)
  • mini-project with presentation
    • data of your interest
    • backup claims with numbers
    • work in teams
  • optional: writing a seminar paper
    • in cooperation with Prof. Sophie Mützel

Organization

  • seminar on Thursday from 2.15pm - 4.00pm
    • additionally, streaming via Zoom
  • course website KED2025 with slides + information
  • readings on OLAT
  • communication on OLAT Forum

Registration via UniPortal

🚨 Registration period: 3 February – 9 March 2025 🚨

Assignment #1 ✍️

  • get/submit via OLAT
    • starting tonight
    • deadline: 28 February 2025, 23:59
  • discuss issues on OLAT forum

Course website

References

Bar-Tal, Omer, Hila Chefer, Omer Tov, Charles Herrmann, Roni Paiss, Shiran Zada, Ariel Ephrat, et al. 2024. “Lumiere: A Space-Time Diffusion Model for Video Generation.” January 23, 2024. https://doi.org/10.48550/arXiv.2401.12945.
Bender, Emily M., Timnit Gebru, Angelina McMillan-Major, and Shmargaret Shmitchell. 2021. “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜.” In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 610–23. Virtual Event Canada: ACM. https://doi.org/10.1145/3442188.3445922.
Brooks, Tim, Bill Peebles, Connor Homes, Will DePue, Yufei Guo, Li Jing, David Schnurr, et al. 2024. “Video Generation Models as World Simulators.” https://openai.com/research/video-generation-models-as-world-simulators.
Chang, Huiwen, Han Zhang, Jarred Barber, A. J. Maschinot, Jose Lezama, Lu Jiang, Ming-Hsuan Yang, et al. 2023. “Muse: Text-To-Image Generation via Masked Generative Transformers.” January 2, 2023. https://doi.org/10.48550/arXiv.2301.00704.
Duquenne, Paul-Ambroise, Brian Ellis, Hady Elsahar, Justin Haaheim, John Hoffman, Hirofumi Inaguma, Christopher Klaiber, et al. 2023. “Multilingual Expressive and Streaming Speech Translation.”
Graham, Shawn, Ian Milligan, and Scott Weingart. 2015. Exploring Big Historical Data: The Historian’s Macroscope. Open Draft Version. Under contract with Imperial College Press. http://themacroscope.org.
Grattafiori, Aaron, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, et al. 2024. “The Llama 3 Herd of Models.” November 23, 2024. https://doi.org/10.48550/arXiv.2407.21783.
Imagen-Team-Google, Jason Baldridge, Jakob Bauer, Mukul Bhutani, Nicole Brichtova, Andrew Bunner, Lluis Castrejon, et al. 2024. “Imagen 3.” December 13, 2024. https://doi.org/10.48550/arXiv.2408.07009.
Karras, Tero, Samuli Laine, Miika Aittala, Janne Hellsten, Jaakko Lehtinen, and Timo Aila. 2020. “Analyzing and Improving the Image Quality of StyleGAN.” March 23, 2020. https://doi.org/10.48550/arXiv.1912.04958.
Lazer, David, Eszter Hargittai, Deen Freelon, Sandra Gonzalez-Bailon, Kevin Munger, Katherine Ognyanova, and Jason Radford. 2021. “Meaningful Measures of Human Society in the Twenty-First Century.” Nature 595 (7866, 7866): 189–96. https://doi.org/10.1038/s41586-021-03660-7.
Lazer, David, Alex Pentland, Lada Adamic, Sinan Aral, Albert-László Barabási, Devon Brewer, Nicholas Christakis, et al. 2009. “Computational Social Science.” Science 323 (5915): 721–23. https://doi.org/10.1126/science.1167742.
Luccioni, Alexandra Sasha, Emma Strubell, and Kate Crawford. 2025. “From Efficiency Gains to Rebound Effects: The Problem of Jevons’ Paradox in AI’s Polarized Environmental Debate.” January 27, 2025. https://doi.org/10.48550/arXiv.2501.16548.
Lundberg, Ian, Jennie E. Brand, and Nanum Jeon. 2022. “Researcher Reasoning Meets Computational Capacity: Machine Learning for Social Science.” Social Science Research 108 (November): 102807. https://doi.org/10.1016/j.ssresearch.2022.102807.
Mollick, Ethan. 2023. “Post-Apocalyptic Education.” September 16, 2023. https://www.oneusefulthing.org/p/post-apocalyptic-education?utm_medium=web.
Radford, Alec, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, and Ilya Sutskever. 2022. “Robust Speech Recognition via Large-Scale Weak Supervision.” December 6, 2022. https://doi.org/10.48550/arXiv.2212.04356.
Salganik, Matthew J. 2017. Bit by Bit: Social Research in the Digital Age. Illustrated edition. Princeton: Princeton University Press. https://www.bitbybitbook.com.
Sheynin, Shelly, Adam Polyak, Uriel Singer, Yuval Kirstain, Amit Zohar, Oron Ashual, Devi Parikh, and Yaniv Taigman. 2023. “Emu Edit: Precise Image Editing via Recognition and Generation Tasks.”
Timiryasov, Inar, and Jean-Loup Tastet. 2023. “Baby Llama: Knowledge Distillation from an Ensemble of Teachers Trained on a Small Dataset with No Performance Penalty.” In Proceedings of the BabyLM Challenge at the 27th Conference on Computational Natural Language Learning, 251–61. Singapore: Association for Computational Linguistics. https://doi.org/10.18653/v1/2023.conll-babylm.24.
Wu, Yuxin, Alexander Kirillov, Francisco Massa, Wan-Yen Lo, and Ross Girshick. 2019. “Detectron2.” Meta Research. https://github.com/facebookresearch/detectron2.